The Evolution of Asset Reliability
Mark Banzin
My experience with equipment maintenance and operational reliability began in 1982 as an avionics technician in the United States Marine Corps, maintaining complex electronic systems on military aircraft. I later transitioned into the maritime and recreational diving industry, where I was responsible for maintaining life-support equipment, compressors, dive vessels and other critical assets. Today, after progressing through operational leadership positions in the water and wastewater industry, I serve as VP of Asset Reliability.
Across these different industries, one challenge has remained consistent: keeping critical equipment operating reliably while minimizing unexpected failures.
Early in my career, maintenance management was largely manual. We relied on equipment logs, inspection records, maintenance schedules and eventually spreadsheets to track performance and anticipate maintenance needs. Despite our best efforts to be proactive, we frequently found ourselves operating reactively, addressing problems after they became apparent.
Computerized maintenance management systems (CMMS) significantly improved our ability to organize maintenance activities, track equipment histories and manage preventive maintenance. Yet much of our strategy remained tied to predetermined schedules rather than actual equipment condition.
Today, AI represents what I believe is the next major evolution in asset reliability.
Building the Foundation: Data before AI
After more than four decades working with critical equipment, I have learned that technology is only as effective as the information supporting it.
Before organizations can successfully implement AI-driven reliability strategies, they must establish accurate asset information, equipment histories, maintenance records, failure classifications and criticality assessments.
In our efforts to strengthen asset reliability across multiple water and wastewater facilities, we have focused on consistent asset identification, standardized maintenance practices and improved work-order management through CMMS technology.
As part of our ongoing asset reliability initiatives, we are integrating artificial intelligence with maintenance management and SCADA systems to improve operational visibility and support more informed maintenance decisions. By bringing maintenance information and operational data together, we are establishing a foundation for more advanced condition-based and predictive maintenance capabilities.
“The future is not about replacing human experience with AI. It is about using AI to make human experience even more valuable.”
Without reliable information, even sophisticated analytical tools can produce misleading recommendations. AI should not replace sound maintenance management practices. It should build upon them.
Connecting Operational Data to Better Decisions
One of the most promising opportunities for AI is connecting maintenance information with real-time operational data.
Traditionally, CMMS platforms document maintenance activities, while supervisory control and data acquisition (SCADA) systems monitor equipment and process performance. Combining these information sources creates opportunities to better understand asset health.
Throughout my years in water and wastewater operations, I have experienced situations where high-pressure pumps in reverse osmosis facilities failed unexpectedly despite established preventive maintenance programs. Looking back, there were often subtle changes in operating conditions that could have indicated developing problems.
Changes in pressure, flow, vibration, or energy consumption may seem insignificant individually, particularly when information is reviewed separately or over extended periods.
This is where I believe AI can make a meaningful difference. By analyzing operating trends alongside maintenance history, AI-assisted systems could identify developing patterns that might otherwise go unnoticed, providing operators with earlier indications of potential equipment problems.
Rather than simply responding to failures or relying exclusively on predetermined maintenance intervals, maintenance teams could make more informed decisions based on actual equipment condition.
Technology Should Strengthen Human Expertise
Despite AI's potential, I firmly believe experienced operators remain among the most valuable assets within any operation.
An experienced operator recognizes subtle changes in equipment behavior that may not immediately appear in a maintenance report. Unusual sounds, vibration, fluctuating process conditions, or changes in system performance often provide early indications that something requires attention.
AI can analyze large volumes of information and identify developing trends, but it cannot replace the practical judgment, situational awareness and operational knowledge developed through years of hands-on experience.
The most effective reliability strategies combine technological capabilities with human expertise. Rather than replacing operators, AI should equip them with better information, allowing them to focus attention where it will have the greatest operational impact.
A Practical Path Toward Predictive Reliability
Organizations considering AI should resist pursuing technology simply because it is available. The starting point should always be identifying the operational problem that needs to be solved.
Whether the objective is reducing unplanned downtime, improving equipment availability, extending asset life, or controlling maintenance costs, successful implementation requires clear objectives and measurable results.
A phased approach allows organizations to establish reliable data, prioritize critical assets, introduce analytical capabilities and evaluate performance before expanding.
Having witnessed the progression from manual maintenance systems to computerized platforms and now artificial intelligence, I believe AI will become an essential component of asset reliability management.
Ultimately, the future of reliability will depend on how effectively we combine reliable information, disciplined maintenance practices, technological innovation and experienced people.
The future is not about replacing human experience with artificial intelligence. It is about using artificial intelligence to make human experience even more valuable.
